farazbabar

Exactly once processing is not possible in distributed systems. Anyone that tries to sell that snake oil is dishonest and anyone who buys it should not be making purchasing decisions. The definition and requirement of idempotent processing means systems must be able to handle messages delivered more than once, which irrefutably proves there is no such thing as exactly once.

Even within centralized, monolithic systems with no outside interaction, laws of physics and reality still apply - a pull on the cord, an earthquake, a flooding or a myriad of other things may interrupt message processing resulting in either exactly zero or more times a message will be processed even if the message were being processed within the confines of an embedded micro-controller using hand crafted assembler code.

Can I please complain a little? Is hackernews the small local news source for the bay area bubble? I don't know what pioneer is or which scores are being manipulated or what happens to winners (or losers). I assumed the electronics company known for its amplifiers etc. was up to some dodgy shenanigans. Zero context from the title. I don't mind clicking stuff to get more context but I think some of it should be there without assuming everyone is in the loop at all time.

I went through it all in less than 2 years and ended up hoarding the things I enjoy very much to this date. I have no regrets (never), coming into millions in thirties is an incredible life experience and the vacations, scotches, watches, cars, homes and other toys are all things I have enjoyed very very much. I also learned to buy forever things which has changed how I see consumption completely. It also led to some amazing opportunities to meet very interesting people in strange places. Overall, I cannot emphasize how much fun it is to spend a few million dollars in 12-18 months without being fixated with savings.

Kind of like CRDT (conflict free replicated data types for the biologists amongst us).

As one grows and evolves, one learns to better themselves. A fairly significant number of smart and capable people have been guilty of acting in not so ideal manner when dealing with incompetent or otherwise incompatible people but the answer to questions like "give me an example of an encounter or interaction that did not go as well as it could" is not to sugarcoat or misrepresent, I think part of growing up is to own one's mistakes honestly, recognize them as such and learn from them and if an interviewer believes they can really ferret out liars that are in reality asshole but have ready answers for such question in order to take a hard pass on folks that disclose past mistakes honestly, well, I am not sure I consider it to be the best strategy, but I am still learning things.

> It must be allowed to discriminate based on ability, even if a certain ability correlates trongly with age, sex or something else.

Wow! I rarely openly disagree with strangers on the internet but wow! I am currently focused on improving diversity and inclusion within my teams and the diversity of thought, approaches and outcomes that are a direct result of this effort are a delight. Progress may not always be in the right direction or slower than one expects but zigging and zagging towards the right thing is a reward in its own.

Aerospike

Multiple criteria:

1. When you go from one machine running the code to more than one 2. Any system that may experience failures and detection of such failures and recovery is desirable 3. Most distributed systems due to the failure scenarios inherent in such systems.

Warning, unpopular opinion coming up. Bay area may have the critical mass for investors and technologists but I think we owe it to ourselves to have some diversity of thought and ambition. I am on the west coast but have avoided moving to bay area due to the lack of intellectual diversity. I know my venues are much more limited in some ways but my experiences and interactions are much richer in others (I know this because I have split my time 50/10/40 between bay area, new york and southern california for nearly a decade since 2010). Weather, access to money and talent are very strong and influential variables but to me, diversity of thought (and approach to capital) are more valuable. Clearly, this is anecdotal, so take it with a rock of salt.

I write about it on my company blog for internal developers, perhaps I can get our legal folks to go over it and I can publish some of it. There is a lot of tribal knowledge for sure but on my journey from issuers to merchants to networks, I keep finding consistent themes. Perhaps it is time to put this all down on paper.

There is a lot more to it than that, there is merchant acquiring (merchant relationship with a payment services provider that gives merchants the credentials and equipment to accept payments), there are sometimes aggregators that transactions then funnel into. Next the transactions hit a network which acts as the router between acquirers and issuers (more on this later). The network can also act as a default stand in when the issuing bank is not reachable. From here the transaction goes into the issuing bank which runs a host of fraud and risk rules to answer the questions: 1. Should I authorize this? <- this is bank's prerogative and mostly deals with financial risk and 2. Can I authorize this? <- this is regulatory in nature and can be very binary and very binding. Its the the second question that also comes with binding regulations in some jurisdictions that prevent all parties along this chain from disclosing the reasons for declining, withholding funds or delay processing to affected parties. Even more interestingly, PayPal operates at most of these tiers almost entirely independently (PayPal API as the acquirer, PayPal as the aggregator, PayPal as the network of network AND of-course PayPal as the issuer with PayPal credit. Obviously not all transactions go through all tiers in case PayPal (a credit card transaction goes to card network and card issuer in the back for example) but Pay upon invoice, Bank transfers and PayPal credit (bill me later) are all examples of PayPal extending credit to buyer. Payments are quite fascinating, you are correct that a lot of issues are political but don't underestimate the complexity of the ecosystem, especially in global context. And I have not even gotten into bad faith actors and the war between good and bad guys yet.

My personal favorite is https://vertx.io because it is a little more transparent in how it works and the Rx abstractions are something I was familiar with when I started using it. Not to mention springboot did not offer reactive programming until fairly recently whereas vertx has been reactive from day 1, rxjava or not.

I will take another look at springboot just to see what the fuss is about. I am a bit surprised that netflix did not come up with their own light weight backend framework based on netty.

In traditional I/O, a hardware interrupt is triggered whenever data arrives at hardware boundary and the interrupt can get serviced by any core that is available to the scheduler. One can imagine how much overhead is involved in context switching whatever that core was doing before, setting up the registers, moving data and then relinquishing the core back to OS - in this model, dedicated cores serve I/O in a memory mapped ring buffer like data structure sized to your application needs. There is no allocation/deallocation overhead, no management beyond moving a pointer and no context switching. If you can spare the cores, this can significantly improve performance.

In one use-case, I was able to quadruple the performance on a 32 core xeon by installing 4 10gbps ethernet cards and dedicating the first eight cores to I/O (2 per interface). This is all about latency but with proper care, it also improves throughput.

The one thing that sets many ultra wealthy, especially the self made ones apart from a lot of people is how much they value time. I especially liked the fictional evolution of ultra wealthy into immortals in "altered carbon" because it alleviated time as the most important and finite thing they have.

UUID is not enough to prevent duplicate resource creation. Consider the following steps: 1. Receive a request with uuid, 2. Persist the fact that system received this request, 3. who or what processes it? Even if you are working off atomic persistence, the resource creation process (the kick-off process) needs to maintain the state that it plucked one off the queue or table, atomically update the state in persistence as it creates the resource somewhere else and log the successful creation of this resource as yet another entry in the persistence store. This way, requests that exceed their SLA can be retried and the downstream systems must act with correct idempotent behavior in order for all of this to work correctly in the presence of failures.

It is all very technically possible, but it goes beyond just a UUID (in other words, UUID is necessary but not sufficient).

This comment is at once so endearing and so alarmingly indicative of the bubble we live.

I think this can be further generalized, that there is a tier of talent that we must excuse from having to filter themselves and that this may serve humanity even if it rubs some the wrong way.

This is already the case with 1099-K and even though the link at https://www.paypal.com/us/smarthelp/article/how-does-paypal-... refers to sales, the catch all mechanism is entirely driven by number of transactions and amounts which are easily attainable with an active social circle.

As someone trying to build a brand new team, I am happy to pay people whatever they think they are worth. I wonder if there is a way to do this without offending the engineers I am trying to attract - I mean, if they are not to answer the question how much do you want, how do I pay them what they want? I am not at all worried about over-paying because any engineers that don't meet the performance bar I set won't last long enough for it to matter.

Modern operating systems are very efficient at scheduling threads with negligible workloads compared to threads that are performing under load. I suspect that a lot of test code for toy examples and default workload for thousands of threads you see on an operating system has running at any given time is probably not cpu intensive enough to matter.

Compared to this, a library designed to perform actual work using hundreds of threads on a modern os with let's say 8 cpu cores - simply won't work due to the constant context switching overhead. Consider the library called LMAX disruptor as an example of this behavior where developers need to configure threads to be equal to number of available cpus if they wish to do busy_wait processing for inbound messages on the ring buffer. Chronicle Q actually uses JNI to pin message processing threads to individual CPU cores to make sure operating system scheduling does not end up flushing L1/L2 caches and that message processing for a given queue continues to occur on the same thread/cpu combo.

So it depends on your use-case and application.

Of course it is. However, divide and rule was the original british mantra. It saddens me personally how well it worked on the Indian sub-continent and as a student of history I wonder why it did.

How sad that British were able to sow enough seeds of hatred and division in India that brothers in arms ended up fighting each other and separating into two not quite at peace neighbors less than a hundred years later.

Oppo makes a great product but the thing that set them apart for me was their ability to play SACD disks. The number of players that can play this format have been dwindling and Oppo's demise is just a sign of the times. I wonder if physical media for very high resolution audio (and video?) would be as far removed a memory as sony beta or if the vinyl sales trends would imply a reversal of fates down the road, vindicating the high resolution video and audiophile crowd.

I don't get the downvotes for this comment. This is a real issue that I have first hand experience with. The poster even qualified the statement saying its probably not the case for this store but it is certainly true of a number of establishments - these operate in the second and third stages of money laundering also known as layering and integration.

Oh boy, that and perturbation.

I have thought about a natural language understanding (NLP/NLU) based startup to do exactly this. So much to do, so little time.

If you paid millions in taxes personally, then you have sufficient funds to qualify for investor/job creator visa (which in turn leads to a green card as well). I have family that were able to accomplish this with as little as 500K and 2 years, but depending on the scenario it can take as much as a couple million dollars (depending on location and nature of business). I am sure this is stuff your attorney must have explained to you so probably there is context I am missing here.

I have been wanting to do this in Phoenix, AZ. We have then ultimate duopoly between cox and century link and no other alternatives for most neighborhoods. I find it perplexing and think that market exists to offer better service. I have some experience launching an ISP back in Pakistan when I was still a teenager but this guide is exactly the kind of motivation I needed to get me going.

As an ex pakistani, I have always admired how much India has been able to overcome the odds and continues to prosper and charge forward. I think the gp has some merit in hoping that within a generation or two, India has a chance to overcome a number of these challenges. I do concur with your assessment of Pakistan however.

I said no to mortgage backed securities in 2006. Beginning to think about targeted advertising disguised as dopamine driven feedback loops along the same lines.